Du Moyen Age jusqu'à la fin du XIXe siècle, le nuage hante le ciel de la peinture occidentale. Moins qu'un motif descriptif, le nuage constitue un élément de la sémiotique picturale, un graphe dont les fonctions varient avec l'époque. A l'origine utilisé, à l'imitation des machines de théâtre, pour faire apparaître le sacré dans le réel (ascension du Christ, visions mystiques), il joue un rôle plus ambigu à la Renaissance, au moment où le modèle perspectif assure la régulation : le nuage vient alors masquer l'irreprésentable infini, en même temps qu'il le désigne, assurant ainsi l'équilibre paradoxal d'une institution picturale intimement liée aux conditions de la science.
Ce qui est finalement tenté ici, c'est, à travers un inventaire des fonctionnements successifs du signifiant "nuage", une redistribution critique des domaines et des rôles assignés à l'art, à la science et à l'idéologie dans une structure de représentation : élément pour restituer à l'histoire de l'art sa dimension systématique et matérialiste.
Hubert Damisch (born 1928), is a French philosopher specialised in aesthetics and art history, and professor at the École des Hautes Études en Sciences Sociales (EHESS) in Paris from 1975 until 1996.
Damisch studied at the Sorbonne with Maurice Merleau-Ponty and, later, with Pierre Francastel. In 1967 he founded the Cercle d’histoire/théorie de l’art that would later become the CEHTA (Centre d'histoire et théorie des arts)[1] at the EHESS.[2]
Damisch has written extensively on the history and theory of painting, architecture, photography, cinema, theatre, and the museum. His works are landmark references for a theory of visual representations.
we are here to attempt a semiological analysis that does not set out by acknowledging its dependence upon the linguistic (phonetic) model, but instead aims to define the specific semiotic function that constitutes the mainspring of pictoral production. such an analysis cannot possibly proceed simply by a functional division of the painted surface into its constitutive parts, and then by breaking down those parts, in their turn, into the elements of which they are composed. on the contrary, it needs to circumvent the flat surface upon which the image is depicted in order to target the image's texture and its depth as a painting, striving to recover the levels, or rather the registers, where superposition (or intermeshing) and regulated interplay--if not entanglement--define the pictorial process in its signifying materiality. however, this should be done without presupposing their relative coherence or the possibility of drawing up a more or less exhaustive list of terms and functions that belong to the same class. in the absence of any explicit theory, the method to be followed must perforce be inductive. it would amount to inferring from an analysis of the pictorial process itself some concept of what we have called its various levels or reigsters, and then describing these, in the physical sense of the tern, and revealing their relative organizational role. (14)
hubert damisch, "sign and symbol" in a theory of /cloud/: toward a history of painting. stanford: stanford up, 2002.
I highly recommend Hubert Damish's A Theory of /Cloud/: Toward a History of Painting. First published in 1972, not translated till 2002, I so wish I had this book when I was taking courses in art history almost 50 years ago.
Dense reading... but if you keep ploughing ahead, understanding becomes cumulative as he returns again and again to the examples on which he builds his argument.
The advantage of reading it now (unless you are able to travel the world to see all the art he cites), is that reading in front of a computer, you can quickly Google images you are not familiar with, or remember only vaguely. Richly documented. THIS is what art history should aspire to.
It's a cliche to saw that a given books form or style mirrors it's content. But that is precisely what happens here. An thoroughly researched an highly intelligent exploration of anti-rationalist themes in the history of Western art. Content-wise, this is everything Panofsky is not.